Prompt management
Prompt management is the systematic creation, organization, versioning, evaluation, and operational control of prompts used in AI applications.
Machine Learning Platform Engineering
Prompt management is the controlled creation, versioning, evaluation, and deployment of prompts for language-model applications.
- How it is used
- Treats LLM prompts as managed production assets rather than hardcoded application strings.
- How deep it goes
- Works through creation, labeling, runtime fetching, compilation, tracking, testing, and caching; does not specify a universal prompt-quality metric.
- Examples
- Uses a documentation assistant’s templated chat prompt, production configuration, tracked generation, and fallback boundary prompt.
- What you can do after
- Apply it via Langfuse to update, deploy, compare, and roll back prompt versions without redeploying application code.
Read how Machine Learning Platform Engineering explains prompt management →
Build AI-Enhanced Web Apps
Prompt management is the organized storage, updating, versioning, and governance of prompts used by an AI application.
- How it is used
- Organizes prompts for dynamic updates and governance in generative AI web applications.
- How deep it goes
- Explains database benefits, versioned prompt files, and mock testing; describes a database-backed architecture without providing its implementation.
- Examples
- Uses versioned translation prompts, a mocked summary test, and a PlantUML diagram-generation library.
- What you can do after
- Evaluate and understand prompt storage, versioning, and testing choices for an AI web application.
Read how Build AI-Enhanced Web Apps explains prompt management →
Prompt Engineering in Practice
A system for managing prompts across their production lifecycle, including collaboration, governance, and lifecycle visibility.
- How it is used
- Keeps production prompts findable, changeable, and accountable as teams and prompt libraries grow.
- How deep it goes
- Details lifecycle steps, PMS capabilities, and audit gaps; leaves deployment pipelines and telemetry backends outside scope.
- Examples
- Contrasts audit-log entries for a properly promoted prompt and a direct production edit investigated after an incident.
- What you can do after
- Practice/perform it by assessing maturity, moving prompts through a lifecycle, and using process signals to prioritize improvements.
Read how Prompt Engineering in Practice explains prompt management →
Generative AI in Action
Prompt management is the systematic design, testing, deployment, refinement, and operational maintenance of prompts for language models.
- How it is used
- Deploys prompts as an orchestration-layer capability for improving language-model responses in enterprise applications.
- How deep it goes
- Explains engineering, optimization, and PromptOps with metrics, feedback, token limits, testing, and version control, but defers fuller governance details.
- Examples
- Uses news summarization, adaptive user-intent prompts, domain-specific language, interaction analysis, A/B testing, and response-quality feedback.
- What you can do after
- Evaluate/understand prompt performance by monitoring accuracy and relevance, refining from feedback, and managing complexity, tokens, versions, and rollbacks.
Read how Generative AI in Action explains prompt management →
Also covered in
- Designing AI SystemsIn early access (MEAP)
- LLM Customization and Fine-TuningIn early access (MEAP)
What to know first
- language modelA language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- productionProduction is the live operational setting in which a deployed software system serves its intended users or processes real workloads.
- experimentationExperimentation is the planned investigation of a question or hypothesis by conducting trials or interventions and observing their outcomes.
- workflowA workflow is an organized sequence of coordinated activities or tasks through which work progresses toward a defined outcome.
Related concepts
- Langfuseimplemented by · Langfuse is an open-source observability platform for tracing, monitoring, analyzing, and evaluating applications built with large language models.
- optimizationhas part · Optimization is the systematic selection or modification of values, configurations, or implementations to improve a defined objective, often subject to constraints.
- prompt engineeringkind of · Prompt engineering is the deliberate design and refinement of inputs, instructions, and context to guide an AI model’s behavior or outputs.
- promptopshas part · PromptOps is the operational discipline of managing prompts and prompt-based workflows through development, evaluation, deployment, monitoring, and maintenance.
- language modeluses · A language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
- maturity journeyused by · The Maturity Journey is a framework for assessing a team's prompt-management maturity and identifying its next practical improvement step.
Questions about prompt management
What is prompt management?
Prompt management is the systematic creation, organization, versioning, evaluation, and operational control of prompts used in AI applications.
Which Manning books explain prompt management?
Machine Learning Platform Engineering, Build AI-Enhanced Web Apps, Prompt Engineering in Practice and Generative AI in Action.
What should I know before learning prompt management?
Manning books that teach prompt management usually introduce language model, production, experimentation and workflow first.
What is prompt management a kind of?
Prompt management is a kind of prompt engineering. Prompt engineering is the deliberate design and refinement of inputs, instructions, and context to guide an AI model’s behavior or outputs.
What are the parts of prompt management?
Prompt management includes optimization and promptops.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.



